All use cases

Upsell & Cross-sell

GLP-1 muscle-preservation supplement upsell

The medication that is working is also the reason your patient needs protein and resistance support — surfaced at the stage that matters, not as a banner.

Get aheadNext best thingBehavioral triggerTime-based triggerChatPushHealth & TelehealthE-commerce & RetailGLP-1 & Peptides

What it does

Rapid weight loss on a GLP-1 medication comes with a known clinical pattern: a meaningful share of what is lost can be lean muscle, and that risk peaks at a predictable program stage rather than on day one. Most programs say nothing about it until a patient asks, and a generic supplement promo would land as spam in a care relationship. The agent waits for the program stage where the risk is clinically relevant for that patient and opens a care-first conversation about protecting muscle, then surfaces the matching protein or supplement option as the answer to a real need the treatment itself created. Doing nothing leaves both a care gap and attach revenue on the table.

How it works

  1. 1
    Trigger. a program-stage milestone fires for a GLP-1 patient (the window where muscle-loss risk is clinically known), optionally reinforced by behavioural signals such as logged rapid loss or reduced activity.
  2. 2
    Decision. the agent checks consent, frequency caps, and memory (program phase, prior offers, dietary preferences, what they already own); a quality judge gates the message so it reads as care, not promotion, and the framing stays inside approved, non-prescriptive language.
  3. 3
    Action. a care-first message goes out on in-app chat or push — what tends to happen at this stage and why protein and resistance support matter — and, only once the need is acknowledged, the matching supplement option appears with a one-line reason and an in-app path to add it.
  4. 4
    Follow-through. the outcome (taken, declined, or not raised) is written to memory; the agent never repeats the offer outside the cap; if the patient reports a clinical concern, the conversation routes to a clinician rather than continuing as a sale.
Example
AgentYou're at the stage of your program where the weight is coming off fastest — this is also when the body can lose muscle alongside fat. Protecting it now keeps your strength and results on track. How are you doing with protein day to day?
CustomerHonestly not great, I'm just not that hungry anymore.
AgentThat's very common on this medication. A high-protein supplement makes it easy to hit the target without forcing big meals — want me to add the one matched to your plan so it's there for next time?

Configuration

How the agent is wired for this use case.

Triggera program_stage_reached time signal from the care program for the muscle-loss-risk window, optionally combined with a rapid_loss_logged behavioural event.
Tools & actions
  • Care program system · read program phase, treatment timeline, and adherence to identify the right stage
  • Customer memory · read dietary preferences, prior offers and declines; write back the outcome
  • Supplement catalogue · select the protein/creatine/supplement option matched to plan and preferences
  • Messaging channel · deliver the care-first opener and render the in-app add path
  • Clinical-safety judge · gate every message against approved, non-prescriptive language and route clinical concerns out
Autonomythe stage-triggered care opener and the attach offer run unattended under judge gating; the agent gives no dosage or medical advice and makes no clinical recommendation — any symptom or treatment question is a mandatory hand-off to a clinician, never an automated reply.
Channelschat · push
Escalationany report of a side effect, a treatment-change request, or distress hands off immediately to a clinician with the conversation context attached.

What you need

The inputs this use case runs on. Your channels stay yours; the agent supplies the judgment.

Signals

program-stage / treatment-timeline events; optional weight-log and activity events that reinforce the risk window.

Data

consent state; program phase and treatment context; customer memory (dietary preferences, prior offers, owned products); supplement catalogue with eligibility rules.

Guardrails

the agent surfaces a product, never a dose or medical instruction — no proactive dosage or clinical advice, with symptom and treatment questions escalated to a clinician; offer only after a need is acknowledged; judge gating on every unprompted message; frequency caps and instant opt-out; product claims confined to approved language.

Metrics it moves

  • attach-rateup: supplements attach to a clinically real, stage-specific need instead of a broadcast promo.
  • adherenceup: protein and resistance support help preserve results, keeping patients on the program.
  • revenue-per-conversationup: care check-ins at this stage carry an attributable, well-matched attach.
  • arpuup: the program monetises a need the treatment itself creates, within the care relationship.

Related use cases

Upsell & Cross-sell

WhatsApp care-opener upsell

the general care-first opener this pins to a clinical stage

Upsell & Cross-sell

Next-best-product recommendation block

the in-reply offer UI, here triggered by physiology not a live question

Retain / Save

Month-4 cliff companion coach (GLP-1)

the retention coach this attach sits alongside

See it on your own customer journey

Bring one drop-off, one churn cliff, or one silent segment. We will show you what a proactive agent with memory and judgment does with it.

Book a demo